Thursday, April 24, 2014

Missing a Train, and Game Design

So here's my first non-review post. It may or may not be interesting.

I've been guilty of not cluing stuff well enough. And I've felt bad about that and tried to rectify it. I feel, to some extent, the responsibility of finding things lies with the player. They need to do a reasonably careful reading, and they should be rewarded for reading carefully to find a hidden clue. All the same, you shouldn't make them sweat it.

I've been on the other side of it and been able to brush this off and, in fact, I'm pretty forgiving if I notice that an author forgot to clue something, but other work they did makes it clear they did so generally. But an experience I had missing a train gave this a more emotional angle. I don't want this to be a shaggy dog story or lengthy complaint but rather an example of how mistakes get in the way of performing simple actions and how things can be exacerbated or dealt with.

The Howard L stop in Chicago is where the Purple, Red and Yellow lines meet. I generally take the Red line and switch to the Purple to visit the Evanston or Wilmette libraries, which are great for checking out DVDs of all kinds of TV series.

One Sunday I tried to get to the Evanston library just before the 6 PM. Generally, the northbound trains take off from the east platform, but one took off from the west without warning. This was sort of confusing, because the screen that tells departure times mentioned a train leaving. So technically I *could* have been looking around to see if a train that maybe should've been on the east track was on the west--a train with the purple "Linden" label. That would be fully rigorous, but it would also be stress inducing an impatient and get in the way of the reading I do.

So when I saw a train pull out with a few people on it--fewer than were waiting on the east platform--I figured I'd just missed something, and while I couldn't make it to Evanston, Wilmette closed at 9. So I had a backup plan. The system had a certain amount of built-in error checking with the departure screens and so forth, even if they were on the concourse adjoining the west and east platforms. So I was able to see what happened, but it required more thought than it should have. And I actually forget if the next train was on the west or east track. I do remember the general confusion, and people wondering if they had a right to be confused.

A week later, re-visiting the library, I exited a purple line train at Howard. There was a red line train on the other side of the platform, but the doors were closed to preserve heat and to let operators perform their inspections. I went under the heat lamps on the other side, since the terminal was crowded, and began reading. I figured I would hear the announcement.

I heard the announcement that the train was about to go south--WITH the doors closed, and with the train about to leave. So I asked someone in the customer service kiosk and was told "You need to pay better attention." I and other people waiting on my side of the platform because it was crowded--as well as people on the train's side, because the train didn't take up the full platform length, and people's view of it were blocked by the stairs up or walled-off stairs down.

My guess was that the customer service person was as confused as I was, so I didn't take their employee number down. I was pretty disappointed, book or no. The next train similarly gave no warnings, and I regret not yelling out to people to board it so they didn't miss it again. The doors didn't stay open for very long, either.

If this were a text adventure, or really, any sort of game, some pretty serious bugs would be at play here.

* inadequate hinting (or discouragement) to the player to perform a clear action they want to (no indication of construction or track switching from the usual: player has to look for obscure )
* hostile or unhelpful parser/error response (the customer service people seemed baffled no employee was available on the platform and shifted the blame to me)

This is mitigated by the general impersonal help of the screens posting arrival times (okay, between annoying ads,) just like a game can sometimes hint something in not the best way, or it can force you to break immersion by hitting an in-game hint menu.

So this experience has given me concrete example and a sort of buzz-phrase: I don't want to make my players miss the train. And I don't want to shift the blame on them if they do, unless they're trying to. It's above "I know there are bugs, and I need to fix them."

It also reenforces how much a designer needs to rely on as much data as he can get. I was the only one who complained, even though others were confused! I think this parallels how I may get a bunch of transcripts where people struggle similarly but can't pull the trigger on needing X or Y, or maybe they think they're being nice by not voicing vague suspicions. To use numbers, if there's a 10% chance someone will complain about something that needs it, you need 7 people just for an even chance it'll get found out.

Fortunately, with transcripts, seeing people getting stuck and missing the point I thought I was making clear is a huge help. Sometimes that's due to me hiding something I coded, and sometimes it's due to me thinking "oh, that'd make the puzzle too easy." And running through my game with a clear mind can't really replicate this, even after I've put a game, or that area of it, aside for a while.

There's a good ending to my train story. I wrote the CTA and informed them that this sort of snafu would be even worse if the temperature dropped (it did. It got VERY cold. Cold enough to push people to the stairwells, out of sight.) They said this was an immediate priority, and the week after, there were clear announcements about trains being on the "wrong" track. And I think most people in a game creation/management position would be very disappointed this sort of thing happens, and they can move to fix it--and if the safety net is at all robust, it's not too bad.

This also encourages me to take a deep breath when I see something missing from a game, where it seems the author left something out. They're not intending to, except in an obvious troll game. And again, an example like this pushes my feelings above "I better not whine--the author didn't mean to" (helpless) to "this instance wasn't intended. Let's see what the author meant."

Of course, even if you're actually testing and not playing the game, having to do too much of this is exhausting, and how much rope the player gives the author is a personal choice. But having a real-world example of something annoying that gets fixed, and fixed well, is heartening to me, from the player or programmer side.

1 comment:

  1. I'm also a big traingoer, and while we don't have winters cold enough in Australia to cause the particular problems you described, I relate to how easily the info systems can start failing you as soon as routines are varied. I think the most telling thing you mentioned for game testing, which a lot of people may not realise, is how you need to increase your sample space of testers to excavate some problems - problems that may have been faced by all the testers but which fell into a 'I'm not sure if I need to speak up' zone.